The overall aim of this project is to identify suspect organic and inorganic chemicals during critical periods of intrauterine exposure that are associated with child developmental disabilities and mental retardation (DD/MR). The study will compare individual level analysis with new spatial and temporal statistical methods to infer from analysis of clusters to individuals about the risk associated with temporal and spatial exposures. The study will be divided into three phases. During Phase I we will test the hypothesis that there are clusters of DD/MR among infants identified by maternal residence during pregnancy. We will link over 152,000 maternal child pairs using Medicaid prenatal and postnatal medical data, and monthly addresses, so clusters of DD/MR can be identified using spatial statistics. During Phase II we will test the hypothesis that soil samples will show variation in toxicity levels that can be used to identify locations with potentially greater concentrations of suspect organic and inorganic chemical which may be associated with DD/MR. This will involve collecting and analyzing soil samples from cluster sites and from comparison areas where rates of DD/MR are not elevated. In addition, we will test the hypothesis that mercury, chromium, arsenic, PCBs, and other selected chemical contaminants will be measured in higher concentration in the higher toxicity soil samples compared to the lower toxicity soils, as identified using Microtox. We will send a subsample of the high and low toxicity soil for chemical analyses at an external analytical laboratory. During Phase III we will test the hypothesis that maternal residence proximal to toxic waste sites with identified chemical exposures, during critical periods of fetal development, will be associated with higher risk for DD/MR in the infant, after controlling for maternal and child risk factors. This will involve the development of statistical methods that combine traditional and spatial procedures to analyze associations between the spatial relationship, temporal exposure (month of pregnancy), and soil samples. These can be used to generate hypotheses about specific environmental exposures associated with DD/MR. This research will be both multidisciplinary and multimethod, with specific emphasis on the development of new techniques, to analyze the complex associations of environmental and medical data.